The aim of this article is to apply advanced predictive modeling techniques to understand the degradation process of microplastics in aquatic environments. Utilizing a Fractional Factorial Central Composite Experimental Plan, this study seeks to develop precise predictive statistical models that enable forecasting the quantity of pollutants generated during the degradation of microplastics under various environmental conditions. This tool was applied to model changes in DOC (dissolved organic carbon) and DEHP (bis(2-ethylhexyl) phthalate) values during the degradation of microplastics in aquatic ecosystems. The methods were developed using data derived from laboratory tests conducted using the GC-MS technique. The obtained approximating functions, considering factors such as degradation time, water temperature, and particle size, significantly reduced the analysis time. A two-stage verification of the approximating functions was conducted, considering the accuracy of the function form, its adequacy, the statistical significance of input variables, and their correlation with DOC and DEHP. The employed a Fractional Factorial Central Composite Experimental Plan allowed for the simultaneous reduction in the number of experiments and prediction of the influence of variables on the output values. Precise predictive models support understanding of the microplastic degradation process, facilitating the development of effective strategies for managing this pollution.